Questions about NIU observations

Hi, I have data from ATUS respondents only, all years, and am trying to understand the NIU (not in universe) coding.

  1. Shouldn’t there be NIU observations for time use variables re: children for those with no children in the household, instead of 0 values for those who don’t have household kids? For example, “bls_carehh_kid” has no NIUs, and the number of observations with value ‘0’ is bigger than the no. of observations with hh_child=0. How can I be sure that the time spent with children variables actually exclude the respondents with no children?

  2. What kind of respondent is NIU in “schlcoll”? There are 87k such observations. How are these different from those who are not enrolled in school or college?

thanks much, Sanjiv

I will aim to address each of your questions one at a time.

(1) If your data only include respondents, then everyone in your data is eligible for all time use questions. One way to ensure that time spent with children excludes respondents with no children is to include the KIDUND18 variable in your data extract. This variable identifies whether the respondent has an own child under the age of 18 living in their household.

(2) As is noted on the universe tab for the SCHLCOLL variable: the universe for this variable includes all ATUS respondents age 15 through 49. So, depending on how you define your data extract (e.g., respondents only, respondents + household members, etc.), anyone with SCHLCOLL==99 “NIU” is either not an ATUS respondent or is less than 15 or older than 49 years old.